Module C7.

Viscous Dissipation Term in

Energy Equations
H. H. Winter
University of Massachusetts
Amherst, Massachusetts.

OBJECTIVES

After completing this module, the student should

be able to:
1. Calculate the rate of viscous dissipation for a
given flow.
2. Calculate the rate of viscous dissipation in a
macroscopic energy balance.
3. Calculate dissipation in a slip layer.
4. Use the dissipation function as a criterion to
distinguish between a viscous and an elastic
material.
5. Calculate dimensionless groups to estimate
the magnitude of viscous dissipation.

for a fluid with suspended particles is equated to the viscous dissipation in a pure Newtonian fluid, both being in
the same flow (same macroscopic velocity gradient).
RATE OF VISCOUS DISSIPATION

The rate at which work is being done on a volume element for changing its volume and its shape is defined as
(for derivation, see Appendix)
u: Vv= -pV

rate of
work for
volume
change

PREREQUISITE MATHEMATICAL SKILLS

1. First year college calculus.

PREREQUISITE ENGINEERING AND
SCIENCE SKILLS
1. First year college physics.

2. Macroscopic balances.

v+~

(1)

rate of work
for shape
change at
constant volume

The stress,

u= -pl+r

(2)

with
1
p= --trace u

INTRODUCTION

Deformation and flow of materials require energy.

This mechanical energy is dissipated, i.e. during the flow
it is converted into internal energy (heat) of the material.
This phenomenon can be demonstrated by performing a
simple experiment with a metal paper clip: bend the clip
wide open and close it repeatedly until the clip breaks.
Now, touch the metal near the region of the break and
feel the high temperature. The mechanical energy for
bending the metal has been converted into internal energy. The increase of internal energy expresses itself in a
temperature rise.
Viscous dissipation is of interest for many applications:
significant temperature rises are observed in polymer
processing flows such as injection molding or extrusion
at high rates. Aerodynamic heating in the thin boundary
layer around high speed aircraft raises the temperature of
the skin. In a completely different application, the dissipation function is used to define the viscosity of dilute
suspensions (Einstein, 1906, 1911): Viscous dissipation
Modular Instruction Series

is divided into the pressure, p, and the extra stress, r.

V v and Vv are the divergence of the velocity vector and
the velocity gradient. The second term in Equation 1 is,
called the "dissipation function,"

c/>=r: Vv

(3)

since most (not necessarily all) of the work is irreversibly

converted into heat. The dissipation function for flows of
Newtonian fluids is given in component form; see Table
1.
VISCOUS DISSIPATION IN PIPE FLOW

The steady flow in a pipe of constant cross section (radius R) will be used in the following for explaining viscous dissipation in bulk and at a slip boundary.
Macroscopic Balance

Flow of a fluid in a pipe requires mechanical energy

which is supplied by a pump or by the hydrostatic pres-

27

Table 1. Components of the dissipation function of a Newtonian fluid with viscosity I'

Cartesian
Coordinates

+" [ -av,ax+avx]
ay-

+J.I

av, avy] 2 [ avx av,] 2

[ -ay+az- +J.I -az+ax-

..
2--1(.V . v) 2]
av, v,) 2+ (av,)
+ (1- -+[( -av,)2
ar
r ae r
az
3
Cylindrical
Coordinates

I a
r ar

1 av, av,
r ae az

v=--(rv,)+--+-

( 1 av, v,)2 + ( - -1 -av~+ -v, +v, -cot-8)

[( -av,)2
ar + --+r ae r
r sin
r r
(J a~

2--I(V . v) 2]
3

Spherical
Coordinates

a ( v~)]
1 av,
+J.I [ - +rr sin -(J aq,
ar -r
V

1 a
r2 fJr

r Sin (J

(4)

The volume flow rate is the same in both cross sections

(assuming constant density)
(5)

const. This gives a total rate of work

(6)

28

av~

v=--(r 2v,)+-.--(v8 sin8)+-.--

sure of a reservoir (potentiai energy). Consider a pipe

segment as shown in Figure 1 and a control volume between cross-sections 1 and 2.
The rate of work done for tlow of a fluid through a
pipe is calculated by integrating the rate of work per unit
surface area, n a v, over the entire surface of the control volume. See also the Appendix Equation 46. Note
that the surface along the pipe wall does not contribute,
since its velocity is zero. The work on cross-sections 1
and 2 can be calculated by assuming uniform pressures
p 1 and p 2 and by neglecting the small influence of the
extra stress T. The rate of work done on the volume of
fluid becomes

where p

a0

r Sin (J fJcp

corrkol volume

L
'(

L
Figure 1. Straight pipe section of length L, diameter D. The con
trot volume includes a thin layer of the stationary wall material.

for deforming, for accelerating, and for elevating the

fluid. In a horizontal pipe of uniform cross-section (Figure 1), the mechanical energy for the pipe flow is completely dissipated, since the kinetic energy (no acceleration) and the potential energy (no change in altitude) of
the fluid do not change between cross-sections 1 and 2.
Note that the macroscopic energy balance gives the total
qissipated energy ("friction loss"), however, no information is given on whether the dissipation is uniform
throughout the volume or whether there are regions of
large viscous dissipation and other regions of negligible
dissipation. An interesting situation arises when the. fluid
American Institute of Chemical Engineers

slips at the wall (see Example 3): one part of the energy
for flow through the pipe is dissipated at the slip surface
and the remaining part is dissipated in the volume of the
deforming fluid.

pressure drop was found to be A.p = 2.34 X 10 5 Pa.

Calculate the temperature increase between inlet and outlet by assuming no conduction through the wall. The relevant properties are:

Example 1: Pipe Flow in Polymer Processing

density p = 1000 kg m- 3

Polymer melt is forced through a pipe of L = 0.04 m

and D = 0.002 m (runner channel to fill the mold of injection molding machine). A pressure drop P1 - p 2 = 8
X 10 7 Pa was typically found to give a volume flow rate
Q = 6 X I0- 6 m 3/s. Calculate the average temperature
increase in the polymer between inlet and outlet. Assume
adiabatic walls, negligible density changes, and a steady
temperature field in the pipe.
Typical values for the physical properties of a polymer
are:

heat capacity Cp=4.2x 10 3 J kg-1 K-1

density p = 10 3 kg/m 3
heat capacity c= 1.4x 10 3 J kg- 1 K- 1
The dissipated energy is calculated from Equation 6:

t:..p

2.34x 10 5

peP

10 3 (4.2 x 10 3 )

t:..(T)=-

5.6x10- 2 K

(11)

Viscous dissipation does not significantly alter the temperature of the water. However, it is still important,
since it determines the power requirement of a pipe line
system, i.e. viscous dissipation determines the size of the
pumps for a pipe system and the energy costs of pumping.

Example 3: Pipe Flow With Slip at Wall

E= Q(p1- P2) = (6 X 10- 6) (8 X 10 7) =4.8 X 10 2 J/s

(7)

For a steady temperature field in a pipe with adiabatic

walls, the entire energy is transported convectively with
the fluid. The convective energy flow through a pipe
cross section is

pc(T)Q=27r [ pcT(r)v(r)r dr

The average temperature increase for adiabatic pipe flow

is calculated as

(8)

The difference between energy convection into the pipe

and out of the pipe is equal to the generation of internal
energy due to viscous dissipation
(9)

The average adiabatic temperature increase between inlet

and outlet is calculated as

Consider a fluid which flows through a straight pipe

section as shown in Figure 1. The fluid is found to slip at
the wall at a velocity VR. Examples of slipping fluids are
highly filled suspensions, linear polyethylene, polybutadiene, and polyvinylchloride in the molten state. Determine how much energy is dissipated in the slip region.
A control volume for a macroscopic energy balance is
chosen around the fluid volume in the pipe section. The
main point is that the control volume does not contain a
layer of stationary wall material, but that the outer surface moves with the finite slip velocity uR. The rate of
work done on the control volume of fluid is again c.alculated by integrating the rate of work per unit surface,
n u v, over the entire surface.
(12)

P1-P2

A.(T) = (T)'L- (T)l=~

8x
10 3

10 7

(1.4 X 10 3 )

57 K

(10)

Local temperatures might by far exceed this average

value. An average temperature increase of 57 K is very
large. Some of the assumptions in the beginning of this
example will have to be reconsidered: the thermal and
the rheological properties can be expected to change significantly between entrance and exit; the high temperatures of the fluid give rise to a substantial radial temperature gradient, i.e. the heat flux into the wall cannot be
neglected anymore.

Example 2: Friction Loss in Pipe Flow of Water

Water at 20oC is flowing through a pipe (L = 30.5
m, D = 0.05 m) at an average velocity of 6.2 m/s. The
Modular Instruction Series

The first part is the rate of work done for flow through
the pipe (see Equation 6) and the second part is (minus)
the rate of work for slip along the pipe wall. The shear
stress at the wall can be expressed in terms of the axial
pressure gradient in the pipe,
(13)

In case of a uniform slip velocity uR and a uniform pressure gradient, ap!az = (P2- p 1)/L, the integral can be
simplified. The rate of work for slip along the wall becomes
(14)
The rate of work for deforming the fluid (subscript d) in

29

the flow through the pipe remains as

r/R.

Ed= Q(pl-P2)-Es

-Na=l
--No= 10

l 1-::!._)
(v)

=(PI-P2)Q (

(15)
Z=O.OI

( v) is the average velocity. The limiting case of plug

flow, vR = (v), obviously requires no energy for the deformation. The other limiting case of no slip, VR -c (v),
requires all the energy to be dissipated in the deforming
fluid.
DISTRIBUTION OF THE DISSIPATION FUNCTION

Most flows are inhomogeneous, i.e. the stress and the

rate of deformation are functions of position. Again,
steady pipe flow is used for demonstrating inhomogeneity.
In pipe flow, the viscous dissipation is not uniform in
the cross-section (see Table 1)
(16)

The shear stress is given by the stress equation of motion,

T

rdp
rz 2 dz

=--

dz

pressure grad'1ent

The shear rate is equal to the radial velocity gradient

.

dVz

(18)

'Yrz=-

iJr

which depends on the type of fluid:

ll.r

JVeW

(vz)

t oman
. fl Ul'd: 'Yrz=.
4
r
--

Power law fluid: i'rz=- (

(19)

R R

r)

1 ) (vz) (
;;+
3 R
R

lin

(20)

The power law viscosity is here defined by

in
(21)

with a reference viscosity of TJ = TJ{i' ) at a reference

shear rate "( 0 in the shear thinning region of the viscosity
plot.
Now, let us determine the distribution of the dissipation function for the power law fluid, keeping in mind
that the solution will include the Newtonian fluid as a
special case (with n = 1). Introducing Equations 20 and
0

30

T-To

f ...-To
Figure 2. Developing temperature profiles' In a pipe with isothermal wall 174). Parameters are the Nahnie number, Na, and the
dimensionless distance from the entrance Z = kll peP { v) R 2 ).
The viscosity is described by a power law with n = 0.4.

21 into the dissipation function cp, Equation 16, gives

)n+l((vz))2(r)l+l!n
-

1
cp(r)=TJ 0 ( -+3
n

(22)

where

(17)

where

-dp = ax1'al

(I)

The dissipation function is zero at the center line and has

its maximum value at the wall.
Developing temperature fields in pipe flow demonstrates the non-uniformity of the viscous dissipation; see
Figure 2. Large radial temperature differences are generated in a fluid which started out being of uniform temperature. These radial temperature differences give rise to
conduction of heat towards the wall. A fully developed
temperature is reached when the heat flow into the wall
balances the viscous heat generation. A method of calculating developing temperatures in many different shear
flow geometries and a review of the literature are given
by Winter (15).
DIMENSIONLESS GROUPS

The calculation of temperature fields requires the solution of the equation of energy. The equation of energy is
conveniently rewritten in dimensionless form. Order of
magnitude arguments allow the elimination of small
terms, as compared to the important terms. This procedure also applies for the viscous dissipation term in the
energy equation. One has to define a dimensionless
group, the generation number Non which indicates the
importance of viscous dissipation as compared to convection, conduction, and compression.
The generation number Non can be defined with the
equation of energy in a most simple form:
(23)

The fluid is assumed to be purely viscous and of constant

viscosity. The thermal properties p, Cp, k are assumed to

American Institute of Chemical Engineers

be constant [an equivalent discussion for the compressible

fluid can be found in Schlichting (13)]. The stress in the
viscous dissipation term is already replaced by

r=rry with i=Vv+(Vv)T

DT oT
Dt at

A Tprocess

(25)

The equation of energy is made dimensionless by scaling it with the factor H 2/kAP. Equation 23 becomes

L Dt*

= V*T* +- - - 71*(1'* : 'Y*)

A T0

T0
71

make the variables dimensionless:

= (T - T0 ) /t:.T0

V*

= HV
= 71/'T/o

71*

i*

= tLIV

= iHIV

The definition of the generation number is

'"

v.,o
Nan= kt:.To

(27)

and its relation to the equation of energy is known. A

large generation number implies that viscous dissipation
cannot be neglected in comparison with heat conduction.
Note that the product 71*i*:i* might locally adopt very
large values ( > 1). Therefore, the viscous dissipation
might locally influence the temperature even if the generation number is smaller than one. A safe value for neglecting the effects of viscous dissipation seems to be
Nan -< 0.1.
The generation number has been defined several ways
and accordingly, has different names in the literature.
This module will discuss two of the most common cases.
The other dimensionless group, the Graetz number, compares the magnitude of convection and conduction. It will
not be discussed here.
Modular Instruction Series

T 0 =A

(30)

Tprocess

V2

NEe Np,=

= reference velocity
= characteristic length in direction of velocity gradient (pipe radius or slit width, for instance)
= characteristic length in flow direction (pipe
length in pipe flow)
reference temperature difference
reference temperature
= reference viscosity, 71( VI H, T0 )

T*
t*

Then this temperature difference is chosen to be the characteristic temperature difference, t:. T 0 ' with which the
temperature changes can be scaled:

(26)

The scaling factors,

(29)

or the product of the Prandtl number, Np, and the Eckert number, NEe (13),

2 kAP

or between the boundaries of the flow,

kATprocess

1 V271o

(28)

v2.,o
NB,=----

Graetz number, N az

Generation number, Nan

= T1- T w

The generation number in this case is called the

Brinkman number (4),

pcpVH H DT*

-----k

There are many flows with a given temperature difference, t:. Tprocess between the fluid and a wall,

(24)

The viscosity has to be specified as a function of temperature and deformation rate. D!Dt is a substantial time
derivative.
- = - + v VT

Flow With a Transverse Temperature Difference

Cp!:. Tprocess

Cp'T/0
--=NBr

Both definitions are equivalent. The name, "Brinkman

number,'' seems to be preferred in studies on developing
temperatures in channel flow and the name, "Eckert
number,'' is preferably used in studies on viscous dissipation in thermal boundary layers.
Flow Without Imposed Temperature Difference

Scaling of the temperature changes due to viscous dis~

sipation becomes more difficult when there is no given
temperature difference t:. Tprocess. This situation always
occurs when isothermal flow conditions are attempted
(which is quite common in polymer processing). Viscous
dissipation would disturb these isothermal conditions and
its extent has to be estimated in modeling efforts.
The most common choice of scaling factor for temperature changes in nearly isothermal processes is

t:. ro = t:. Trheol = - ( a.,;aT) To,-yo

(33)

The temperature changes are of interest since they affect

the viscosity and hence, affect the flow pattern. A viscosity with a temperature dependence of the Arrhenius type,

.,=aebiT

(34)

where T =absolute temperature

gives a characteristic rheological temperature difference,

!::.. Trheol = b

where T0 =absolute reference temperature

For molten polymers,

t:.Trheol

= 30-70 K
31

The temperature differences are then scaled as (T T0 ) /~ Trheol and the generation number is called the
Nahme number (11).

(36)

In nearly isothermal processes, the Nahme number is a

measure of how much viscous dissipation affects the temperature dependent viscosity. Large values of NNa indicate that isothermal conditions cannot be maintained.
Note that the temperature change and the generation
number cannot be scaled with a temperature level T0
For scaling, one always has to use a temperature difference. A quantity (T - T0 )/T0 ) would be dimensionless;
however, it would adopt different values in different temperature scales (Fahrenheit, Celsius, Kelvin).

Figure 3. Sandwich device for shearing a material.

iment, Equation 38:

.
OUx OUy U0
'Yxy=-+-=iJy ax h

COS

(wt)

(40)

The stress in the Hookean material. is proportional to the

shear strain,
(41)

Txy= G')'xy

ENERGY DISSIPATION AND STORAGE

The dissipation function becomes

The work done for deforming a viscous material is irreversibly converted into internal energy, i.e. it is dissipated. The work done for deforming an elastic material,
however, is stored as potential energy, which can be recovered mechanically. An example is a. rubber band in
the stretched state. It can perform work when contracting
to its original length. The dissipation function, T:Vv, is
applicable to both cases. It gives the rate of work done
for deforming a material, independent of whether this
material is viscous, elastic, or viscoelastic. The dissipation function is always positive when applied to viscous
materials. It adopts positive or negative values with elastic and with viscoelastic materials. The name "dissipation
function'' is actually misleading when describing storage
and recovery phenomena in deforming elastic materials.

Example 4: Oscillatory Strain of a Hookean Material

An elastic material is placed between two extensive
parallel plates, as shown in Figure 3. The lower plate is
stationary and the upper plate moves with velocity.
(37)

.
.
G
Txy'Yxy = G')'xy'Yxy = w

(uh )2 .
0

sm (wt) cos (wt)

(uh )2 .

=G
-

2w

sm (2wt)

(42)

One cycle requires work

W=

h/w .
G
jo Txy'Yxy dt=2w

G(v )2 ( -1)

=2w

2w

( u0
h

2 j27r/w sm. (2wt) dt

o

1w=0
cos (2wt)jh
0
'

(43)

As expected for the elastic material, the work per cycle

is equal to zero. The work done in one half of a cycle is
recovered in the other half.
The same experiment, however, with a Newtonian
fluid, would be described by a dissipation function:

Txy"(xy=JJ.("(xy) 2 =JJ.

(~ )

2
COS

(wt)

(44)

aty = h
and work per cycle
A momentum balance gives the velocity in between the
plates. For uniform properties, it has the simple form

ux(y)=h

U0

cos (wt)

W=

r2r/w

Jo

(38)

T :

Vv = Txy"Yxy

(39)

The shear rate is given by the velocity field of the exper-

32

hU)2 cos 2 (wt) dt

0

U)2
( h 7r
0

=JJ.
Calculate the rate of work done in order to deform the
material. Integrate to find the total work for one cycle.
Compare the result to the behavior of a Newtonian fluid
in the same experiment.
The rate of work done is given by the dissipation function,

JJ.

(45)

The work is dissipated in the material.

APPENDIX: DERIVATION OF THE

DISSIPATION TERM
Consider a small volume element (volume V, surface
S) of a material which is deformed by a stress a on the
surface. The rate of work done on the surface of the material element is calculated by integrating (n a v)s over

American Institute of Chemical Engineers

the entire surface,

E=

LITERATURE CITED

H(n u v)s dS

(46)

1.

n is the outward unit normal on the surface and (n u)s is

the force per unit area of surface. Multiplication with the
surface velocity gives the rate of work per unit surface.
The work done on the surface is used for deforming
the material element (change of shape and volume), accelerating it, and increasing its potential energy. This is
shown in the following. It will lead to an expression for
the rate at which work is being done for deforming the
volume element.
The surface integral of Equation 46 is replaced by a
volume integral (using the Gauss theorem):

(n u v)s dS=

Hj

(47)

Vuv=u:Vv+vVu

(48)

The physical meaning of the product v V u can be explained from the stress equation of motion. Scalar multiplication of the local velocity v with the stress equation
of motion (3),

pD
- - u 2 =v V u+pv g
2Dt

(49)

gives an expression for vVu. D!Dt is the substantial

time derivative. The last two equations are introduced
into the volume integral, Equation 47. The result of the
derivation is

ij (n u v)s dS

4.
5.
6.
7.
8.

10.
11.
12.
13.
14.
15.

REFERENCES FOR FURTHER READING

Equation of Energy:
Bird, Stewart and Lightfoot, 1960
Astarita and Marrucci, 1974
Eckert and Drake, 1972

Dimensionless Groups:
Armstrong and Winter, 1982
Pearson, 1978
Schlichting and Kestin, 1955
Eckert and Drake, 1972
Winter, 1977

Polymer Processing:

[ u:

3.

9.

V u v dV

and the kernel in volume integral is replaced by the identity

2.

Armstrong, R. C., and H. H. Winter, "Heat Transfer for NonNewtonian Fluids," in "Heat Exchanges Design and Data Book,"
Section 2.5.12, E. U. Schliinder Ed., Hemisphere Pub!. London
(1982).
Astarita, G., and G. Marrucci, "Principles of Non-Newtonian
Fluid Mechanics," McGraw Hill, London (1974).
Bird, R. B., Stewart, W. E., and E. N. Lightfoot, "Transport
Phenomena," Wiley, New York (1960).
Brinkman, H. C., Appl. Sci. Research, A2, 120-124 (1951).
Cox, H. W., and C. W. Macosko, A/ChE J., 20, 785-795
(1974).
Dinh, S. M., and R. C. Armstrong, A/ChE J, 28, 294-301
(1982).
.
.
Eckert, E. R. G., and R. M. Dilike, "Analysis of Heat Transfer," McGraw Hill, London (1972).
Ei.nstein, A., Ann. Phys., 19, 286 (1906); Ann. Phys., 34, 591
(1911).
Gavis, J., and R. L. Laurence, Ind. Eng. Chern. Fund., 7, 525527 (1%8).
Landau, L. D., and E. M. Lifshitz, "Fluid Mechanics," Pergamon Press, Oxford, (1959).
Nahme, R., Ing-Archiv, JJ, 191-209 (1940).
Pearson, J. R. A., Polym. Eng. Sci., 18, 222-229 (1978).
Schlichting, H., and J. Kestin, "Boundary Layer Theory," McGraw Hill, London (1955).
Winter, H. H., Polym. Eng. Sci., 15, 84-89 (1975).
Winter, H. H., Adv. Heat Transfer, 13, 205-267 (1977).

Vv+~ D

2Dt

u2 -pv

g] dV

(50)

Rate of change of
potential energy
per unit volume.

Thermal Boundary Layer:

Rate of change of
kinetic energy per
unit volume.

Heat Transfer Coefficient for FLOW WITH VISCOUS DISSIPATION:

,__-------Rate of work for

changing the volume
and shape, per unit
volume (see Eq. 1).
u: Vv is the dissipation term in the equation of energy.
For many applications in polymer processing, the
changes in kinetic and potential energy are negligibly
small; all the work done on the surface is practically used
for deforming the volume elem~nt.

Modular Instruction Series

Winter 1977, 1975

Pearson, 1978
Cox and Macosko, 1974
Gavis and Laurence, 1968
Dinh and Armstrong, 1982
EckertandDrake, 1972
Schlichting and Kestin, 1955

Eckert and Drake, 1972

Winter, 1977

Suspension Viscosity Defined with DISSIPATION

FUNCTION:
Einstein, 1906, 1911
Landau and Lifshitz, 1959

STUDY PROBLEMS
1. Calculate the dimensionless temperature (T - T0 )/T0
in degrees Fahrenheit, Celsius, and Kelvin. Use T =

33

with fluid I (viscosity p.1) and the upper half with fluid

n (viscosity P.n). The viscosities differ by a factor of

10: p.1 = lOp.n. Where is the rate of viscous dissipation higher, in the viscous fluid I or in the less viscous fluid ll?

HOMEWORK PROBLEMS
Figure 4. Shear flow of two Immiscible fluids In a gap between
parallel plates.

180C and T0 = 150C. Discuss how a dimensionless temperature can depend on the choice of temperature scale. HINT: Each temperature scale refers to a
different temperature as zero temperature.
2. State the system of equations for the velocity, the
temperature, and the viscosity for flow in a pipe with
isothermal walls. How are the equations coupled with
each other?
3. When is the dissipation function positive and when is
it negative?
4. Consider shear flow of two immiscible Newtonian fluids in a narrow gap between two parallel plates, see
Figure 4. The flow is due to the parallel movement of
the upper plate. The lower half of the gap is filled

34

1. Calculate the Nahme number for pipe flow of Example 1. Use ATrheol = 50 K and k = 0.2 W/mK.
2. Use slip data from the literature to determine viscous
dissipation in the slip layer.
Ref.: L. L. Blyler and A. C. Hart, "Capillary Flow
Instability of Ethylene Polymer Melts,'' Polym. Eng.
Sci., 10, 193-203 (1970).
3. Extend Example 4 to a linear viscoelastic material
with a shear stress

Hint: Determine the time dependent shear stress Txy(t)

first.
4. Calculate the rate of viscous dissipation cf>(r, 8) in a
Newtonian fluid which flows around a single sphere
(see Reference 3, p. 133).

American Institute of Chemical Engineers

Series C:

TRANSPORT
Volume 7:

Calculation and
Measurement Techniques
for Momentum, Energy and
Mass Transfer
R. J. Gordon, Series Editor

AMERICAN INSTITUTE OF CHEMICAL ENGINEERS

 Copyright 1987
American Institute of Chemical Engineers
345 East 47th Street, New York, N.Y. 10017

The appearance of the code below indicates the copyright owner's

consent that, for a stated fee per article, copies of articles in this serial may
be made for personal or internal use, or for the personal or internal use of
specific clients. This consent is given on the condition that the copier pay
the per-copy fee for each article (appearing as part of the code) through the
Copyright Clearance Center, Inc., 21 Congress Street, Salem, Mass. 01970,
for copying beyond that permitted by Sections 107 or 108 of the U.S.
Copyright Law. This consent does not extend to copying for general
distribution, for advertising or promotional purposes, for inclusion in a
publication, or for resale.

LC 80-25573
ISSN 0270-7632/87/$3.75
ISBN 0-8169-0415-4

INTRODUCTION

In 1975 a new venture in education by and for the chemical engineering community was
initiated. Prepared by the CACHE Corporation (Computer Aids for Chemical Engineering
Education) and under the sponsorship of the National Science Foundation (Grant HES 7503911), a series of small self-study fundamental concept modules for various areas of
chemical engineering were commissioned, Chemical Engineering Modular Instruction,
CHEMI.
It has been found in recent studies that modular study is more effective than traditional
instruction in both university and continuing education settings. This is due in large measure to the discrete focus of each module, which allows the student to tailor the speed and
order of his or her study. In addition, since the modules have different authors, each writing
in his or her area of special expertise, they can be produced more quickly, and students may
be asured of timely information. Finally, these modules have been tested in the classroom
prior to their publication.
The educational effect of modular study is to reduce, in general, the number of hours
required to teach a given subject; it is expected that the decreased time and expense involved in engineering education, when aided by modular instruction, will attract a larger
number of students to engineering, including those who have not traditionally chosen engineering. For the practicing engineer, the modules are intended to enhance or broaden the
skills he or she has already acquired, and to make available new fields of expertise.
The modules were designed with a variety of applications in mind; They may be pursued
in a number of contexts: as outside study, special projects, entire university courses (credit
or non-credit), review courses, or correspondence courses; and they may be studied in a
variety of modes: as supplements to course work, as independent study, in continuing
education programs, and in the traditional student/teacher mode.
A module was defined as a self-contained set ofleaming materials tat covers one or more
topics. It should be sufficiently detailed that im outside evaluation could identify its educational objectives and determine a student's achievement of these objectives. A module
should have the educational equivalent of a one to three hour lecture.
The CHEMI Project Staff included:
E. J. Henley, University of Houston, Director
W. Heenan, Texas A & I University, Assistant Director

Steering Committee:
L. B. Evans, Massachusetts Institute of Technology
G. J. Powers, Carnegie-Mellon University
E. J. Henley, University ofHouston
D. M. Himmelblau, University of Texas at Austin
D. A. Mellichamp, University of California at Santa Barbara
R. E. C. Weaver, Tulane University

Editors:
Process Control: T. F. Edgar, University of Texas at Austin
Stagewise and Mass Transfer Operations: E. J. Henley, University of
Houston, J. M. Calo, Brown University
Transport: R. J. Gordon, University of Flordia
Thermodynamics: B. M. Goodwin, Northeastern University
Kinetics: B. L. Crynes, Oklahoma State University
H. S. Fogler, University of Michigan
Material and Energy Balances: D. M. Himmelblau, University of Texas
at Austin

iv

American Institute of Chemical Engineers

CONTENTS

INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

iv

CONVERTING THE EQUATION OF MOTION TO DIMENSIONLESS

FORM ........................................... W. F.. Beckwith

NEWTONIAN FLOW THROUGH FITTINGS, BENDS, CONTRACTIONS,

EXPANSIONS AND NON-CIRCULAR DUCTS ........... K. A. Solen

C7.3

VISCOSITIES OF NON-NEWTONIAN FLUIDS ..... Leon Y. Sadler, Ill

19

C7.4

VISCOUS DISSIPATION TERM IN ENERGY EQUATIONS

................................................... H. H. Winter

27

GENERAL ONE-DIMENSIONAL STEADY-STATE DIFFUSION

PROBLEMS ...................................... Charles E. Glatz

35

C7 .6

COUPLED TRANSPORT ....... Dorothy Lozowski and Pieter Stroeve

41

C7. 7

APPLICATION OF ELECTROCHEMICAL LIMITING CURRENT

TECHNIQUE TO THE STUDY OF INTERFACIAL MASS
TRANSFER -INTRODUCTION AND THEORY
............. Robert F. Savinell, Frank W. Klink ,and John R. Sauter

47

APPLICATION OF ELECTROCHEMICAL LIMITING CURRENT

TECHNIQUE TO THE STUDY OF INTERFACIAL MASS
TRANSFER- EXAMPLES OF APPLICATIONS
............................ Robert F. Savinell and Frank W. Klink

55

APPENDIX: SOLUTIONS TO THE STUDY PROBLEMS. . . . . . . . . . . . . . . . . . . . . . . . .

63

C7. 1
C7 .2

C7.5

C7 .8

Solutions to the Homework Problems are available as a separate reprint

from the AIChE Educational Services Dept., 345 East 46th St., New York,
NY 10017. The cost is $5.00.

Curriculum Analysis: E. J. Henley, University of Houston

The second phase of the project, designed to fill in gaps as well as develop new modules,
is under the direction of D. M. Himmelblau, University of Texas at Austin.

Steering Committee:
B. Carnahan, University of Michigan
D. E. Griffith, Oklahoma State University
L. Harrisberger, University of Alabama
D. M. Himmelblau, University of Texas at Austin
V. Slamecka, Georgia Institute of Technology
R. Tinker, Technology Education Research Center

Editors(* indicates a new task force head):

Process Control: T. F. Edgar, University of Texas at Austin
Stagewise and Mass Transfer Operations: J. M. Calo, Brown University
E. J. Henley, University of Houston
Transport: R. J. Gordon, University of Florida
Thermodynamics: G. A. Mansoori*, University ofllinois at Chicago Circle
Kinetics: B. L. Crynes, Oklahoma State University
H. S. Fogler, University of Michigan
Material and Energy Balances: E. H. Snider*, University of Tulsa
Design of Equipment: J. R. Beckman, Arizona State University
Volume 1 of each series will appear in 1980; Volume 2 in 1981; and so forth. A tentative
outline of all volumes to be produced in this series follows:

SERIES C: TRANSPORT
Volume 1. Momentum Transport'and Fluid Flow

Cl.l
Cl.2
Cl.3
C 1.4
Cl.S
Cl.6

Simplified One-Dimensional Momentum Transport Problems

Friction Factor
Applications of the Steady-State Mechanical Energy Balance
Flow Meters
Packed Beds and Fluidization
Multi-Phase Flow

G. K. Patterson
R. J. Gordon and N. H. Chen
D. W. Hubbard
W. F. Beckwith
W. J. Hatcher, Jr.
R. A. Greenkorn and D. P. Kessler

Volume 2. Momentum Transport, VIscoelasticity and Turbulence

C2.1
C2.2
C2.3
C2.4
C2.5

Non-Newtonian Flow !-Characterization of Fluid Behavior

Non-Newtonian Flow li-Fully Developed Tube Flow
Viscoelastic Fluid Flow Phenomena
Turbulence: General Aspects Illustrated by Channel or Pipe Flow
Turbulent Drag Reduction

D.
D.
D.
N.

V. Boger and A. L. Halmos

V. Boger and A. L. Halmos
V. Boger and R. I. Tanner
S. Berman
G. K. Patterson

Volume 3. Equation of Motion, Boundary Layer Theory and Measurement Techniques

C3.1
C3.2
C3.3
C3.4
C3.5
C3.6

Measurements of Local Fluid Velocities

Equation of Motion
Navier Stokes Equation for Steady One-Directional Flow
Boundary Layer Theory

Boundary Layer Theory: Approximate Solution Techniques

Diffusivity Measurement Techniques in Liquids

N.
G.
G.
R.
R.
V.

S. Berman and H. Usui

K. Patterson

C. April
J. Gordon
L. Cerro
L. Vilker

Volume 4. Mathematical Techniques and Energy Transport

C4.1
C4.2
C4.3

Mathematical Techniques !-Separation of Variables

Mathematical Techniques ll-Combination of Variables
Elementary Steady-State Heat Conduction

Modular Instruction Series

R. S. Subramanian

R. S. Subramanian
W. J. Hatcher

C4.6

Natural Convection
Unsteady-State Heat Conduction
Differential Energy Balance

R. D. Noble
K. 1. Hayakawa
R. D. Noble

CS.I
CS.2
CS.3

Unsteady-State Diffusion
Mass Transfer in Laminar Flow
Turbulent Mass Transfer

S. Uchida
S. H. Ibrahim
S. H. Ibrahim

C4.4

C4.S

Volume 6. Transport Phenomena-Special Topics

C6.1
C6.2
C6.3
C6.4

C6.S
C6.6

T. G. Theofanous
R. S. Subramanian
R. H. Notter
R. H. Notter
P. Stroeve
R. Srinivasan and P. Stroeve

Bubble Dynamics: An Illustration of Dynamically Coupled Rate Processes

Miscible Dispersion
Biomedical Examples of Transport Phenomena 1-Coupled Diffusion Effects
Biomedical Examples of Transport Phenomena 11-Facilitated Diffusion
Mass Transfer in Heterogeneous Media
Advancing Front Theory

Volume 7. Calculation and Measurement Techniques for Momentum, Energy and Mass Transfer

C7.l
C7.2
C7.3
C7.4
C7.5
C7.6
C7.7
C7.8

Converting the Equation of Motion to Dimensionless Form

Newtonian Flow Through Fittings, Bends, Contractions, Expansions and Non-Circular Ducts
Viscosities of Non-Newtonian Fluids
Viscous Dissipation Term in Energy Equations
General One-Dimensional Steady-State Diffusion Problems
Coupled Transport
Application of Electrochemical Limiting Current Technique to the Study of Interfacial
Mass Transfer-Introduction and Theory
Application of Electrochemical Limiting Current Technique to the Study of Interfacial
Mass Transfe~-Examples Q( Applications

W. F. Beckwith
K. A. Solen
L. Y. Sadler, ill
H. H. Winter
C. E. Gratz
D. Lozowski and P. Stroeve
R. F. Savinell, F. W. Klink
and J. R. Sauter
R. F. Savinell and F. W. Klink

Publication and dissemination of these modules is under the direction of Harold I. Abramson, Staff Director, Educational
Activities, AIChE. Technical Editor is Lori S. Roth. Chemical engineers in industry or academia who are interested in
submitting modules for publication should direct them to H. I. Abramson, Staff Director, Educational Activities, American
Institute of Chemical Engineers, 345 East 47th Street, New York, N.Y. 10017.

vi

American Institute of Chemical Engineers